Lines Matching refs:rcx
144 // rcx = -(length to move), a multiple of 64 less than 2GB
149 movq %rdx,%rcx // copy length
153 andq $-64,%rcx // get number of bytes we will copy in inner loop
158 addq %rcx,%rsi // point to 1st byte not copied
159 addq %rcx,%rdi
161 negq %rcx // now generate offset to 1st byte to be copied
211 addq %rcx,%rsi // restore ptrs to 1st byte of source and dest
212 addq %rcx,%rdi
225 cmpl $(-kFastUCode),%ecx // %rcx == -length, where (length < kVeryLong)
230 movdqa (%rsi,%rcx),%xmm0
231 movdqa 16(%rsi,%rcx),%xmm1
232 movdqa 32(%rsi,%rcx),%xmm2
233 movdqa 48(%rsi,%rcx),%xmm3
235 movdqa %xmm0,(%rdi,%rcx)
236 movdqa %xmm1,16(%rdi,%rcx)
237 movdqa %xmm2,32(%rdi,%rcx)
238 movdqa %xmm3,48(%rdi,%rcx)
240 addq $64,%rcx
249 movdqa -1(%rsi,%rcx),%xmm0 // prime the loop by loading 1st quadword
251 movdqa 15(%rsi,%rcx),%xmm1
252 movdqa 31(%rsi,%rcx),%xmm2
253 movdqa 47(%rsi,%rcx),%xmm3
254 movdqa 63(%rsi,%rcx),%xmm4
264 movdqa %xmm1,(%rdi,%rcx)
265 movdqa %xmm2,16(%rdi,%rcx)
266 movdqa %xmm3,32(%rdi,%rcx)
267 movdqa %xmm4,48(%rdi,%rcx)
269 addq $64,%rcx
278 movdqa -2(%rsi,%rcx),%xmm0 // prime the loop by loading 1st source dq
280 movdqa 14(%rsi,%rcx),%xmm1
281 movdqa 30(%rsi,%rcx),%xmm2
282 movdqa 46(%rsi,%rcx),%xmm3
283 movdqa 62(%rsi,%rcx),%xmm4
293 movdqa %xmm1,(%rdi,%rcx)
294 movdqa %xmm2,16(%rdi,%rcx)
295 movdqa %xmm3,32(%rdi,%rcx)
296 movdqa %xmm4,48(%rdi,%rcx)
298 addq $64,%rcx
307 movdqa -3(%rsi,%rcx),%xmm0 // prime the loop by loading 1st source dq
309 movdqa 13(%rsi,%rcx),%xmm1
310 movdqa 29(%rsi,%rcx),%xmm2
311 movdqa 45(%rsi,%rcx),%xmm3
312 movdqa 61(%rsi,%rcx),%xmm4
322 movdqa %xmm1,(%rdi,%rcx)
323 movdqa %xmm2,16(%rdi,%rcx)
324 movdqa %xmm3,32(%rdi,%rcx)
325 movdqa %xmm4,48(%rdi,%rcx)
327 addq $64,%rcx
337 movaps -4(%rsi,%rcx),%xmm0 // 4-byte aligned: prime the loop
341 movaps 12(%rsi,%rcx),%xmm1
342 movaps 28(%rsi,%rcx),%xmm2
345 movaps 44(%rsi,%rcx),%xmm3
348 movaps 60(%rsi,%rcx),%xmm4
352 movaps %xmm0,(%rdi,%rcx)
355 movaps %xmm1,16(%rdi,%rcx)
356 movaps %xmm2,32(%rdi,%rcx)
358 movaps %xmm3,48(%rdi,%rcx)
360 addq $64,%rcx
369 movdqa -5(%rsi,%rcx),%xmm0 // prime the loop by loading 1st source dq
371 movdqa 11(%rsi,%rcx),%xmm1
372 movdqa 27(%rsi,%rcx),%xmm2
373 movdqa 43(%rsi,%rcx),%xmm3
374 movdqa 59(%rsi,%rcx),%xmm4
384 movdqa %xmm1,(%rdi,%rcx)
385 movdqa %xmm2,16(%rdi,%rcx)
386 movdqa %xmm3,32(%rdi,%rcx)
387 movdqa %xmm4,48(%rdi,%rcx)
389 addq $64,%rcx
398 movdqa -6(%rsi,%rcx),%xmm0 // prime the loop by loading 1st source dq
400 movdqa 10(%rsi,%rcx),%xmm1
401 movdqa 26(%rsi,%rcx),%xmm2
402 movdqa 42(%rsi,%rcx),%xmm3
403 movdqa 58(%rsi,%rcx),%xmm4
413 movdqa %xmm1,(%rdi,%rcx)
414 movdqa %xmm2,16(%rdi,%rcx)
415 movdqa %xmm3,32(%rdi,%rcx)
416 movdqa %xmm4,48(%rdi,%rcx)
418 addq $64,%rcx
427 movdqa -7(%rsi,%rcx),%xmm0 // prime the loop by loading 1st source dq
429 movdqa 9(%rsi,%rcx),%xmm1
430 movdqa 25(%rsi,%rcx),%xmm2
431 movdqa 41(%rsi,%rcx),%xmm3
432 movdqa 57(%rsi,%rcx),%xmm4
442 movdqa %xmm1,(%rdi,%rcx)
443 movdqa %xmm2,16(%rdi,%rcx)
444 movdqa %xmm3,32(%rdi,%rcx)
445 movdqa %xmm4,48(%rdi,%rcx)
447 addq $64,%rcx
457 cmpl $(-kFastUCode),%ecx // %rcx == -length, where (length < kVeryLong)
459 movapd -8(%rsi,%rcx),%xmm0 // 8-byte aligned: prime the loop
463 movapd 8(%rsi,%rcx),%xmm1
464 movapd 24(%rsi,%rcx),%xmm2
466 movapd 40(%rsi,%rcx),%xmm3
468 movapd 56(%rsi,%rcx),%xmm4
471 movapd %xmm0,(%rdi,%rcx)
473 movapd %xmm1,16(%rdi,%rcx)
474 movapd %xmm2,32(%rdi,%rcx)
476 movapd %xmm3,48(%rdi,%rcx)
478 addq $64,%rcx
487 movdqa -9(%rsi,%rcx),%xmm0 // prime the loop by loading 1st source dq
489 movdqa 7(%rsi,%rcx),%xmm1
490 movdqa 23(%rsi,%rcx),%xmm2
491 movdqa 39(%rsi,%rcx),%xmm3
492 movdqa 55(%rsi,%rcx),%xmm4
502 movdqa %xmm1,(%rdi,%rcx)
503 movdqa %xmm2,16(%rdi,%rcx)
504 movdqa %xmm3,32(%rdi,%rcx)
505 movdqa %xmm4,48(%rdi,%rcx)
507 addq $64,%rcx
516 movdqa -10(%rsi,%rcx),%xmm0 // prime the loop by loading 1st source dq
518 movdqa 6(%rsi,%rcx),%xmm1
519 movdqa 22(%rsi,%rcx),%xmm2
520 movdqa 38(%rsi,%rcx),%xmm3
521 movdqa 54(%rsi,%rcx),%xmm4
531 movdqa %xmm1,(%rdi,%rcx)
532 movdqa %xmm2,16(%rdi,%rcx)
533 movdqa %xmm3,32(%rdi,%rcx)
534 movdqa %xmm4,48(%rdi,%rcx)
536 addq $64,%rcx
545 movdqa -11(%rsi,%rcx),%xmm0 // prime the loop by loading 1st source dq
547 movdqa 5(%rsi,%rcx),%xmm1
548 movdqa 21(%rsi,%rcx),%xmm2
549 movdqa 37(%rsi,%rcx),%xmm3
550 movdqa 53(%rsi,%rcx),%xmm4
560 movdqa %xmm1,(%rdi,%rcx)
561 movdqa %xmm2,16(%rdi,%rcx)
562 movdqa %xmm3,32(%rdi,%rcx)
563 movdqa %xmm4,48(%rdi,%rcx)
565 addq $64,%rcx
575 movss (%rsi,%rcx),%xmm0 // prefetch 1st four bytes of source, right justified
579 pshufd $(0x93),4(%rsi,%rcx),%xmm1 // load and rotate right 12 bytes (mask -- 10 01 00 11)
580 pshufd $(0x93),20(%rsi,%rcx),%xmm2
581 pshufd $(0x93),36(%rsi,%rcx),%xmm3
582 pshufd $(0x93),52(%rsi,%rcx),%xmm4
590 movaps %xmm1,(%rdi,%rcx)
591 movaps %xmm2,16(%rdi,%rcx)
593 movaps %xmm3,32(%rdi,%rcx)
594 movaps %xmm4,48(%rdi,%rcx)
596 addq $64,%rcx
605 movdqa -13(%rsi,%rcx),%xmm0 // prime the loop by loading 1st source dq
607 movdqa 3(%rsi,%rcx),%xmm1
608 movdqa 19(%rsi,%rcx),%xmm2
609 movdqa 35(%rsi,%rcx),%xmm3
610 movdqa 51(%rsi,%rcx),%xmm4
620 movdqa %xmm1,(%rdi,%rcx)
621 movdqa %xmm2,16(%rdi,%rcx)
622 movdqa %xmm3,32(%rdi,%rcx)
623 movdqa %xmm4,48(%rdi,%rcx)
625 addq $64,%rcx
634 movdqa -14(%rsi,%rcx),%xmm0 // prime the loop by loading 1st source dq
636 movdqa 2(%rsi,%rcx),%xmm1
637 movdqa 18(%rsi,%rcx),%xmm2
638 movdqa 34(%rsi,%rcx),%xmm3
639 movdqa 50(%rsi,%rcx),%xmm4
649 movdqa %xmm1,(%rdi,%rcx)
650 movdqa %xmm2,16(%rdi,%rcx)
651 movdqa %xmm3,32(%rdi,%rcx)
652 movdqa %xmm4,48(%rdi,%rcx)
654 addq $64,%rcx
663 movdqa -15(%rsi,%rcx),%xmm0 // prime the loop by loading 1st source dq
665 movdqa 1(%rsi,%rcx),%xmm1
666 movdqa 17(%rsi,%rcx),%xmm2
667 movdqa 33(%rsi,%rcx),%xmm3
668 movdqa 49(%rsi,%rcx),%xmm4
678 movdqa %xmm1,(%rdi,%rcx)
679 movdqa %xmm2,16(%rdi,%rcx)
680 movdqa %xmm3,32(%rdi,%rcx)
681 movdqa %xmm4,48(%rdi,%rcx)
683 addq $64,%rcx
741 subq %rcx,%rdx // adjust length
753 movq %rdx,%rcx // copy length
755 andq $-64,%rcx // get number of bytes we will copy in inner loop
756 subq %rcx,%rsi // point to endpoint of copy
757 subq %rcx,%rdi
762 movdqa -16(%rsi,%rcx),%xmm0
763 movdqa -32(%rsi,%rcx),%xmm1
764 movdqa -48(%rsi,%rcx),%xmm2
765 movdqa -64(%rsi,%rcx),%xmm3
767 movdqa %xmm0,-16(%rdi,%rcx)
768 movdqa %xmm1,-32(%rdi,%rcx)
769 movdqa %xmm2,-48(%rdi,%rcx)
770 movdqa %xmm3,-64(%rdi,%rcx)
772 subq $64,%rcx
781 movdqu -16(%rsi,%rcx),%xmm0
782 movdqu -32(%rsi,%rcx),%xmm1
783 movdqu -48(%rsi,%rcx),%xmm2
784 movdqu -64(%rsi,%rcx),%xmm3
786 movdqa %xmm0,-16(%rdi,%rcx)
787 movdqa %xmm1,-32(%rdi,%rcx)
788 movdqa %xmm2,-48(%rdi,%rcx)
789 movdqa %xmm3,-64(%rdi,%rcx)
791 subq $64,%rcx